Machine learning,artificial neural networks and social research

Quality & Quantity - Machine learning (ML), and particularly algorithms based on artificial neural networks (ANNs), constitute a field of research lying at the intersection of different...     

Predicting the geo-temporal variations of crime and disorder

Traditional police boundaries—precincts, patrol districts, etc.—often fail to reflect the true distribution of criminal activity and thus do little to assist in the optimal allocation of police resources. This paper introduces methods for crime incident forecasting by focusing upon geographical areas of concern that transcend traditional policing boundaries. The computerised procedure utilises a geographical crime incidence-scanning algorithm to identify clusters with relatively high levels of crime (hot spots). These clusters provide sufficient data for training artificial neural networks (ANNs) capable of modelling trends within them. The approach to ANN specification and estimation is enhanced by application of a novel and noteworthy approach, the Gamma test (GT).     

基于Echo State Neural Networks的短期交通流预测算法

在城市交通环境,交通流的正确预测是比较困难,因为多个十字路口,这使得预置的交通控制模型之间的相互作用和intertwinement不能保持始终高性能在所有的交通情况。     

Advanced manufacturing systems: supply–demand matching of manufacturing resource based on complex networks and Internet of Things

ABSTRACT

After investigation on the existing advanced manufacturing systems (AMSs), it is found that supply–demand matching of manufacturing resource is one of the common issues to be addressed in all AMSs, and methods for addressing this issue have evolved from P2P (peer-to-peer)-based, to information centre-based, and to platform (or system)-based matching, and are moving towards socialisation and service-based solutions. In order to adapt to this trend, a new method for manufacturing resource supply–demand matching based on complex networks and Internet of Things (IoT) is proposed, and a four-layered architecture for implementing this method is designed. In this method, IoT technology is employed to realise the intelligent perception and accessing of various manufacturing resources and capabilities (MR&C), which enables logical aggregation of various distributed MR&C in the form of services. Then complex networks model and theory are used to realise the efficient manufacturing service management, optimal-allocation, and supply–demand matching. In this article, the specific key technologies for implementing the method are presented, including key technologies for manufacturing service generation and aggregation, manufacturing demand/task management, supply–demand matching of MR&C in the form of services, and value/utility adding based on manufacturing service network (MSN), manufacturing task network (MTN) and manufacturing enterprises collaborative network (ECN).     

Quality analysis of a categorical response with different important considerations

Most quality improvement or quality analysis frequently focused on the issue of the quantitative quality response. The issue of addressing a qualitative or a categorical quality response is seldom mentioned. Until now, only a few studies addressed the parameter optimization for achieving quality improvement for a categorical response. However, the weight effect for different categorical level of response cannot be included into their analysis and it will limit the rationality and feasibility for the real applications. The objective of this study is to propose a procedure about quality improvement based on artificial neural networks (ANNs) technique to deal with the parameter optimization of categorical response with different weight effect. A case study involving a taping process from a lead frame (L/F) manufacturer in Taiwan’s science-based park demonstrates the rationality and feasibility of the proposed approach.     

Models of representation of social mobility and inequality systems. A neural network approach

This paper reports on the results of the application of an innovative technique, i.e. neural network models, to mobility data. Our primary aim is to show that the technique is more flexible than traditional statistical modeling, and that it entails less strong methodological assumptions concerning the phenomenon which they are intended to represent. Two kinds of networks have been applied: heteroassociative networks, used for prevision and class membership recognition; and autoassociative networks, used for simulation tasks. Results obtained from experiments with neural networks on Italian data are highly consistent with the body of knowledge derived from previous classical analysis. The explicative power of neural network models proved to be higher than that of path analysis given their capacity to uncover any kind or relation between variables, whether linear or nonlinear. When compared to log-linear models, they enable the reconstruction of mobility processes within a global frame, controlling all relevant variables at once.     

Forecasting CPI inflation components with Hierarchical Recurrent Neural Networks

We present a hierarchical architecture based on recurrent neural networks for predicting disaggregated inflation components of the Consumer Price Index (CPI). While the majority of existing research is focused on predicting headline inflation, many economic and financial institutions are interested in its partial disaggregated components. To this end, we developed the novel Hierarchical Recurrent Neural Network (HRNN) model, which utilizes information from higher levels in the CPI hierarchy to improve predictions at the more volatile lower levels. Based on a large dataset from the US CPI-U index, our evaluations indicate that the HRNN model significantly outperforms a vast array of well-known inflation prediction baselines. Our methodology and results provide additional forecasting measures and possibilities to policy and market makers on sectoral and component-specific price changes.     

电力企业信息安全风险评估模型研究

以信息安全管理标准ISO17799的规定为基础构建了电力企业信息安全风险评估指标体系,运用熵权法确定评价指标权重,并以此作为BP神经网络输入的初始权重,提出了BP神经网络信息安全风险评估模型,仿真结果显示,评价结果是令人满意的。     

边坡影响因素的自组织神经网络归类处理

文章根据自组织神经网络的基本原理,结合55个边坡实例,应用matlab进行编程,建立了边坡影响因素分类处理的神经网络模型,并运用该模型对不同的边坡进行了分类,分类结果提高了神经网络的边坡指标数据的学习效率,从而证明了自组织神经网络对提高用于预测边坡稳定性神经网络性能的有效性。     

A new perspective on the competitiveness of nations

The capability of firms to survive and to have a competitive advantage in global markets depends on, amongst other things, the efficiency of public institutions, the excellence of educational, health and communications infrastructures, as well as on the political and economic stability of their home country. The measurement of competitiveness and strategy development is thus an important issue for policy-makers. Despite many attempts to provide objectivity in the development of measures of national competitiveness, there are inherently subjective judgments that involve, for example, how data sets are aggregated and importance weights are applied. Generally, either equal weighting is assumed in calculating a final index, or subjective weights are specified. The same problem also occurs in the subjective assignment of countries to different clusters. Developed as such, the value of these type indices may be questioned by users. The aim of this paper is to explore methodological transparency as a viable solution to problems created by existing aggregated indices. For this purpose, a methodology composed of three steps is proposed. To start, a hierarchical clustering analysis is used to assign countries to appropriate clusters. In current methods, country clustering is generally based on GDP. However, we suggest that GDP alone is insufficient for purposes of country clustering. In the proposed methodology, 178 criteria are used for this purpose. Next, relationships between the criteria and classification of the countries are determined using artificial neural networks (ANNs). ANN provides an objective method for determining the attribute/criteria weights, which are, for the most part, subjectively specified in existing methods. Finally, in our third step, the countries of interest are ranked based on weights generated in the previous step. Beyond the ranking of countries, the proposed methodology can also be used to identify those attributes that a given country should focus on in order to improve its position relative to other countries, i.e., to transition from its current cluster to the next higher one.     

基于BP神经网络的物流量预测方法研究

人工神经网络作为重要的综合评价方法,能够解决众多非线性问题,把它运用到地区货运量的预测中,可以解决因数据资料缺失,预测结果与真实结果相差较大的问题,为物流业发展提供重要的数据参考。经过实例分析证明基于神经网络的物流量预测模型是行之有效的。     

Efficient organization of information processing

The paper examines the application of the concept of economic efficiency to organizational issues of collective information processing in decision making. Information processing is modeled in the framework of the dynamic parallel processing model of associative computation with an endogenous setup cost of the processors. The model is extended to include the specific features of collective information processing in the team of decision makers which may lead to an error in data analysis. In such a model, the conditions for efficient organization of information processing are defined and the architecture of the efficient structures is considered. We show that specific features of collective decision making procedures require a broader framework for judging organizational efficiency than has traditionally been adopted. In particular, and contrary to the results available in economic literature, we show that there is no unique architecture for efficient information processing structures, but a number of various efficient forms. The results indicate that technological progress resulting in faster data processing (ceteris paribus) will lead to more regular information processing structures. However, if the relative cost of the delay in data analysis increases significantly, less regular structures could be efficient. Copyright © 2007 John Wiley & Sons, Ltd.     

Networks,weak signals and technological innovations among SMEs in the land-based transportation equipment sector

On apprend plus par la conversation des Doctes,

que par la lecture de leurs livres

Les épistres de Seneque

Translation by François de Malherbe,

Paris, Anthoine de Sommaville, 1639, p. 21

Small and medium-sized enterprises, because of their limited resources, use a variety of sources and are linked to different networks to obtain the information they need to develop their strategy and then to gradually organize their environment. Among other things, networks keep them up-to-date with changes in the economy and allow them to take advantage of opportunities to innovate, thus remaining ahead of their competitors. The networks – personal or business – with which these firms interact the most are usually geographically or sociologically close by, embedded in the environment, and are known as strong tie networks. They generally supply signals in a familiar language, based on habit as well a good reciprocal knowledge, which are easy to understand. In addition to this, however, the most dynamic firms also have contacts with weak tie networks, which are further removed from the usual behaviours of entrepreneurs and provide weak signals that, while difficult to grasp and decode, nevertheless offer new, pre-competitive information that can support major innovations. Very little empirical research has been done so far to test the probability of this theory. This paper reports on the results of a survey involving 147 SMEs, all in the land-based transportation equipment sector. It confirms the importance of weak tie networks as opposed to other types of networks, recognizing their complementary contribution to technological innovation. The organization's absorptive capacity is also found to be a significant intermediary factor in taking advantage of weak tie networks.     

基于神经网络集成的失业预警方法

提出采用神经网络集成技术对中国失业预警系统进行建模,以克服当前失业预警系统建模中存在的小样本、高维度、非线性、噪音数据等难题。采用BP神经网络回归模型对失业率进行预测;基于两种集成技术Bagging与AdaBoost对多个神经网络进行集成,以获得比单个预测模型更好的精度与稳定性;最后基于广东省的社会经济调查数据进行了实证分析,实验结果表明:在对失业率的预测上,Bagging集成方法的预测效果优于Adaboost集成方法,也优于单个最好的神经网络模型。     

基于BP神经网络的工程造价估测方法

本文把信息扩散原理和神经网络相结合,提出一种工程造价的估测方法,并给出计算实例。     

Mixed-frequency machine learning: Nowcasting and backcasting weekly initial claims with daily internet search volume data

We propose an out-of-sample prediction approach that combines unrestricted mixed-data sampling with machine learning (mixed-frequency machine learning, MFML). We use the MFML approach to generate a sequence of nowcasts and backcasts of weekly unemployment insurance initial claims based on a rich trove of daily Google Trends search volume data for terms related to unemployment. The predictions are based on linear models estimated via the LASSO and elastic net, nonlinear models based on artificial neural networks, and ensembles of linear and nonlinear models. Nowcasts and backcasts of weekly initial claims based on models that incorporate the information in the daily Google Trends search volume data substantially outperform those based on models that ignore the information. Predictive accuracy increases as the nowcasts and backcasts include more recent daily Google Trends data. The relevance of daily Google Trends data for predicting weekly initial claims is strongly linked to the COVID-19 crisis.     

Secure smart grid communications and information integration based on digital watermarking in wireless sensor networks

As more and more wireless sensor nodes and networks are employed to acquire and transmit the state information of power equipment in smart grid, we are in urgent need of some viable security solutions to ensure secure smart grid communications. Conventional information security solutions, such as encryption/decryption, digital signature and so forth, are not applicable to wireless sensor networks in smart grid any longer, where bulk messages need to be exchanged continuously. The reason is that these cryptographic solutions will account for a large portion of the extremely limited resources on sensor nodes. In this article, a security solution based on digital watermarking is adopted to achieve the secure communications for wireless sensor networks in smart grid by data and entity authentications at a low cost of operation. Our solution consists of a secure framework of digital watermarking, and two digital watermarking algorithms based on alternating electric current and time window, respectively. Both watermarking algorithms are composed of watermark generation, embedding and detection. The simulation experiments are provided to verify the correctness and practicability of our watermarking algorithms. Additionally, a new cloud-based architecture for the information integration of smart grid is proposed on the basis of our security solutions.     

The concept of individual semantic maps in clinical psychology: a feasibility study on a new paradigm

In this paper we propose a new technology able to map the underlying connection scheme among several psychological variables in a single individual. Nine patients with chronic heart failure underwent at regular intervals, two electronic questionnaires to evaluate depression (STAI—short form) and anxiety (STAY-6). Individual semantic maps were developed by Auto Contractive Map, a new data mining tool based on an artificial neural networks acting on the small data set formed by questionnaires items applied serially along time. The clinical psychologist involved in the clinical evaluation of the cases was asked to score the consistency between the information emerging from the graph depicting the structure of the main relationships among items and the clinical picture resulting from the psychological colloquium. All cases reported overall judgments of a good consistency suggesting that the mathematical architecture of the system is able to capture in the dynamics of items value variations through time the underlying construct of the patient psychological status. This technology is promising in remote monitoring of patients’ psychological condition in different settings with the possibility to implement personalized psychological interventions.     

Integration of hybrid wireless networks in cloud services oriented enterprise information systems

This article presents a hybrid wireless network integration scheme in cloud services-based enterprise information systems (EISs). With the emerging hybrid wireless networks and cloud computing technologies, it is necessary to develop a scheme that can seamlessly integrate these new technologies into existing EISs. By combining the hybrid wireless networks and computing in EIS, a new framework is proposed, which includes frontend layer, middle layer and backend layers connected to IP EISs. Based on a collaborative architecture, cloud services management framework and process diagram are presented. As a key feature, the proposed approach integrates access control functionalities within the hybrid framework that provide users with filtered views on available cloud services based on cloud service access requirements and user security credentials. In future work, we will implement the proposed framework over SwanMesh platform by integrating the UPnP standard into an enterprise information system.